Recent scientific and technological developments have made genome-wide association studies a reality. Their success in disentangling the genetic basis of complex diseases will depend largely on the efficient handling of the statistical challenges posed by such studies. Several hundred thousand SNPs will be genotyped and examined for the potential associations with phenotypes. Genome-wide association studies must translate the markedly increased amount of SNP-information into increased statistical power. For the number of statistical tests computed in a genome-wide association study, standard statistical methods for handling the multiple testing problem, such as false-discovery rate, are too conservative and are likely to dilute any true genetic signals. Novel statistical methodology is required to handle the multiple testing problems at this scale. We will develop novel statistical methodology to solve these major hurdles in genome-wide association studies. Our novel statistical methodology will enable researchers to examine the underlying genetic mechanism of complex diseases, such ad Alzheimer Disease and ADHD. PUBLIC HEALTH RELEVANCE: Alzheimer Disease and ADHD are major public health problems in the United States. Our novel statistical methodology will provide a set of tools, which researchers and clinicians can use to identify factors (inherited or environmentally-induced) that affect the development of these diseases. In turn, a better understanding of the genetic mechanisms of these conditions can result in better and more efficient care of those who suffer from - or are most at risk for the development of -these diseases.